ListeningMind Data API · US, South Korea, Japan
See what people search before and after any keyword
Get the real order of searches around a keyword: what came before it and what came next. Keyword APIs stop at volume.
1 Lima = $0.077. Errors are never billed.
No need to pick a tool
Answered with what people in Korea, Japan and the US actually search for: right away when a stored analysis fits, otherwise after you see the price.
For marketers
This page describes the Data API. Marketers can run analyses right away in the apps below with one member sign-in. Every app uses the same ListeningMind search data.
The moments people think of your category, their tastes, what is rising now, a map of search intent
Where the search journey leaks, brand rankings, demand for the next six months, brand names, Naver Smart Store, an ad for each buying situation
Make your brand part of AI answers: prompts, pages, a brand wiki, product pages and AI visibility
Costs are shown in Lima (1 Lima = $0.077), with an estimate before each run.
Search behavior data · Korea, Japan, US
See what people who are not your customers yet search for, and why they don't buy
From what people in Korea, Japan and the US actually search, we extract buying moments, choice criteria, goals, the reasons they can't buy and the reasons they avoid it, and deliver it as data. These are answers a survey struggles to get.
Agents use the same data over MCP and the API, and people use it right here.
Sample results are free, no key needed. Live data calls need a key and a monthly budget cap.
Connect an agent
{
"mcpServers": {
"listeningmind": {
"url": "https://daas.listeningmind.ai/mcp"
}
}
}Paste this into Claude, ChatGPT or Cursor (Cline: add "type": "streamableHttp"). Free tools work without a key; for paid tools, call request_key and have a person approve the link.
curl -X POST https://daas.listeningmind.ai/v1/keyword_info \
-H "Authorization: Bearer $LM_API_KEY" \
-H "Content-Type: application/json" \
-d '{"keywords":["running shoes"],"market":"us","max_lima":2}'The OpenAPI contract is at /openapi.json. Errors come as problem+json with a next field that says what to do.
// A browser agent with this site open
await document.modelContext.callTool("get_sample", {
tool: "keyword_info", market: "us"
});Every page registers catalog, estimate_cost and get_sample; tool pages also register that tool. Works only in browsers that support it.
GET https://daas.listeningmind.ai/.well-known/agent-card.json
POST https://daas.listeningmind.ai/a2a
{"jsonrpc":"2.0","id":1,"method":"message/send",
"params":{"message":{"role":"user","parts":[{"kind":"text",
"text":"Find the buying moments for sunscreen in Japan"}]}}}Hand over a task and the Concierge agent picks the tools, the expected cost and sample results. On staging it plans only.
Why ListeningMind
Every keyword API returns volume, CPC and related keywords. ListeningMind returns those too, and adds the order people searched in. That order shows what they compared, what worried them and where they went next.
For "robot vacuum"
| Monthly search volume | volume |
| Cost per click | cpc |
| Ad competition | competition |
| Related keywords | keyword list |
You know the keyword is popular. You don't know what people searched next, what they compared, or which needs led them there.
For "robot vacuum" · Illustrative
It also returns volume, CPC and competition, so for demand research you don't need a second keyword API.
Each example is the highest-volume node per element in the stored 'running shoes' run. More topics in the topic library.
Who uses it
Language models can't tell you what people searched last month or what they searched next. Let your agent call ListeningMind and cite the result.
Your user asks"What do people compare before buying a robot vacuum?"
For US food, supplement, beauty and appliance makers, and for brands from the UK, Canada, Australia or Europe planning a US launch. Load category paths and buying situations into your BI tool.
Your team asks"Which needs lead US shoppers to creatine, and what do they search next?"
Run the same question with gl=kr or gl=jp. Korea volume includes Naver, the search engine most US tools leave out.
Your team asks"How do Korean shoppers search for sunscreen compared with US shoppers?"
Endpoints
POST JSON to https://listeningmind-data-api.ascentlab.io with your key in the LM-API-KEY header. Set gl to us, kr or jp, one market per request.
Tools
Every tool has the same name over REST, MCP and WebMCP. Data tools are charged for the data a call really returns, at Data API rates. See the estimate with estimate_cost first; send max_lima and the charge never goes above it.
intent_elements70 Lima per topic · search free · first 2 freeThe 15 elements of a stored topic, with evidence. Read right away, no new build.
keyword_info13 Lima per 100 keywordsMonthly volume trend, competition and CPC, and searcher gender and age (Korea)
intent_finder1 Lima per call + 26 Lima per 1,000 resultsRelated searches for a keyword, ordered by volume
path_finder1 Lima per call + 520 Lima per 1,000 keywords in the pathsThe searches people actually made before and after this one
cluster_finder2 Lima per call + 650 Lima per 1,000 keywords in the clustersSearches grouped into intent clusters: how consumers frame the category
ai_overview2 Lima per keywordTop results, the AI overview text and the sources it cites
trend_finderComing soonSearches whose volume grew fastest within a period
Free, no key · catalog estimate_cost get_sample request_key
Coverage
| Data | United Statesgl=us | South Koreagl=kr | Japangl=jp |
|---|---|---|---|
| Search volume, CPC, competition | Google + Naverreported separately and combined | ||
| Search paths and buying situations | Yes | Yes | Yes |
| Searcher gender and age | Not available | Yes | Not available |
| Daily trend index | Yes | Yes | Yes |
| Search results pages and AI Overviews | Yes | Yes | Yes |
Naver handles a large share of searches in South Korea, so Google-only data gives an incomplete picture there. ListeningMind reports Naver volume next to Google.
Data sources
Your legal and data teams will ask. Here is the short version. The full methodology is available on request.
Google search data in all three markets, plus Naver in South Korea. Includes the last 3-month average, 12-month total and month-by-month volume.
Groups of searches linked by search paths and related-search relationships, labeled with purchase context and Category Entry Points.
Built daily from Google autocomplete and search results since May 26, 2025. Each issue links to its sources.
Stored search result snapshots from the last 12 months. Each item includes the date it was collected.
All results are aggregated search data.
Integration
Call the endpoints from your backend, pipelines or product.
POST JSON to listeningmind-data-api.ascentlab.ioLM-API-KEY headerAdd ListeningMind to ChatGPT, Claude, Copilot or Amazon Bedrock. Analysts ask in plain language and the tool picks the endpoint.
Load data into your warehouse on a schedule. Large results arrive as JSON files with a SHA-256 checksum.
Pricing
1 Lima = $0.077 (VAT excluded). Data is measured in credits and paid in Lima: 1,000 credits = 13 Lima = $1. Every response reports the credits that call used. Prices are set in Korean won and shown in US dollars at KRW 1,300 = $1.
Up to 1,000 Lima a month, right away
1,000 credits = 13 Lima
$1 · no monthly fee, no setup cost
Above 1,000 Lima a month
$1,923 a month
500K credits included each month · no setup cost
For long, heavy use
Quote after a talk
Priced to your scope · then $1 per 1,000 credits, no monthly fee
For subscribers who look at data while chatting in Claude or ChatGPT. Lima comes with the plan (from $600 a month, listeningmind.com price), and each call returns a limited result.
Take the data whole for automation, in-house agents and your own services. No subscription; you pay for the keywords you receive. A report is usually 10 to 100 Lima, about the price of one agent run.
Buying from us
FAQ
Those APIs are built around single keywords: volume, CPC and rankings. ListeningMind returns those metrics and adds search paths, the searches people make before and after a keyword, and groups them into buying situations. It also covers Naver in South Korea.
Each Data API endpoint has a set price in Lima (1 Lima = $0.077). One keyword with 300 path keywords costs 157 Lima ($12.08: 1 Lima per call + 300 × 520 Lima per 1,000). Metrics for 1,000 keywords cost 130 Lima ($10: 1,000 × 13 Lima per 100). Set a monthly cap and calls stop at your budget; heavy use gets a lower unit price.
No. 4xx and 5xx responses are not billed. If a call succeeds but returns no results, only the per-call price is charged.
Keyword metrics include the last 3-month average, the 12-month total and monthly volume. Search paths and buying situations are available for the current period and for 3, 6, 9 and 12 months ago. The trend index is updated daily.
Searcher gender and age are available for South Korea. US and Japan results do not include demographics.
Yes, with the right agreement. Internal analysis is allowed on every plan.
Use the ListeningMind web app. Plans include a 7-day free trial and the same search path and buying situation data in the browser.
We suggest the endpoints you need, estimate your monthly Lima, and prepare a test with your own keywords.